Background (Why smoking ?)
Column
Smoking rank as a second leading cause of death worldwide.
Relational Data Model
Column
Relational Data Model
Missing Values
Column
As our data indicates, 49.3% of the data is missing. This is due to the fact that when we attempt to join tables, not all variables have data from the same year. As we expand our data during the process of joining and creating the fact table, missing values become evident in the fact table. Since our goal is to observe and analyze trends, these missing values are OK.
Missing counts in each feature
Column
Missing Values
Type of data
Gender Trend
Column
This Gender differences over time may relate to cultural, physiological, and behavioral factors.
Gender Differences
Column
Smoking prevalanve and mortality rate both high in male than female.Over the years there is positive trend of declining both factors for both male and female.
smoking Prevalence (%)
Column
smoking prevalence
Price in USD for Cigaratte
Column
Price in USD for cigaratte
Europe
Column
Countries with higher tobacco prices tends to have low smoking prevalence, reduced mortality rate and longer life expectancy. However this trend doesn’t hold true for all the countries.
Tobacco prevalence/Price
Column
Correlation between price and life expectancy
correlation (3D plot)
Column
If we look closer, for some countries High tobacco price coexist with high tobacco use. And Some nations with high smoking prevalence have low mortality rate and vice-versa. There may be some confounding factors play role.. such as GDP, Healthcare quality.
In this plot- bigger the size of bubble , lower the tobacco prevalence
NZ Trends (Julia)
Column
Changes in cost of Cigarattes, smoking prevalence,mortality and life expectancy in NZ
Column
changes in cost of cigarattes, smoking prevalence and mortality rate in NZ
Extra plot
Column
Price /prevalence/Mortality trends
Column
Tobacco prevalence /Gender
Readable Name(Fact table)
Column
The data obtained from the WHO was bit difficult to read due to the complex indicator names. To address this issue, we have modified the indicator names to make them more easily readable. This file has been created to display the actual name of the data, along with its description and the newly revised, more readable name.